Publishes metrics to NT using a protobuf under
`photonvision/coprocessors/metrics` using the device host name as the
key.
Refactors metrics to use numbers where possible, instead of strings.
Removes GPU mem display from metrics card when it can't be determined.
Updates UI metrics periodically.
Resolves#1988Closes#830
---------
Co-authored-by: Matt <matthew.morley.ca@gmail.com>
Co-authored-by: Gold856 <117957790+Gold856@users.noreply.github.com>
This PR implements several refinements to the way that NetworkManager
controls the network interface.
- The monitor detects and logs changes to the network address
- The monitor detects and logs changes to the connection and will
reinitialize the connection if needed
- Remove NetworkInterface.java class, which wasn't used anywhere
- Use java.net.NetworkInterface to get IP addresses for any interface
(device)
- Adds a metric for the current IP address (address on the currently
selected interface)
This PR changes the way that photonvision interacts with nmcli to control networking on the coprocessor. Instead of modifying an existing connection, Photonvision adds new connections for DHCP and Static IP configurations. It then activiates the proper one at startup and any time that the network configuration is changed.
It also now uses the interface name and not the connection name and checks that the interface is available before making any changes. If the saved interface is not found, it updates the stored interface name and applies the network settings to the current interface. This should minimize the failure to control the network if the network interface wasn't available when PhotonVision first booted.
One other benefit of not altering the default configuration is that, if PhotonVision fails to run for any reason, the device can be accessed using the original networking configuration.
The code has been tested on an OrangePi5 and and a Raspberry Pi 4.
Addresses: #1261
This PR is for part 1 of #1354. It focuses on adding a model selection
interface for models that exist in `photonvision_config/models/`. Upon
completion we can ship more than 1 model and users could upload their
own through `ssh` without deleting the shipped model. This PR also adds
the abstractions need to support more DNN backends (say OpenCV, or RPI
AI Kit)
Up next is adding a CRUD interface for managing models through the UI.
Our current code matches cameras in this order (which I think is objectively wrong and stupid)
- by-id (/dev/v4l/by-id/product-string)
- by path (/dev/videoN)
- product string/name, but ascii only
- asks cscore to reconnect to cameras using `path`, which on linux is actually /dev/videoN. This isn't guaranteed to stick to a camera if you replug them weirdly at runtime.
This is silly and does not consider the actual physical usb port. I propose instead, in this order:
- By physical usb port path and base name
- by physical usb port path and USB VID/PID
- By base name only (with a toggle switch to disable this, and create a new VisionModule instead)
- Give cscore /dev/video/by-path on Linux systems, pinning Photon USBCameras to a particular usb port once created.
This changes lots of things so stay paranoid!
Tested on Orange Pi 5 and Cool Pi 4B. Merge with parts of the OpenCV DNN PR.
Adds support for YOLOv5s models for Rockchip CPUs with a NPU. Right now hard coded to a note model from alex_idk. Very much still incubating and largely untested.
Allows logging software and live data view to see results. Also removes the requirement for AScope to keep up with the packet serde schema and instead just use the Protobuf descriptor.
